[xemacs-hg @ 2001-09-20 06:28:42 by ben] The great integral types renaming. The purpose of this is to rationalize the names used for various integral types, so that they match their intended uses and follow consist conventions, and eliminate types that were not semantically different from each other. The conventions are: -- All integral types that measure quantities of anything are signed. Some people disagree vociferously with this, but their arguments are mostly theoretical, and are vastly outweighed by the practical headaches of mixing signed and unsigned values, and more importantly by the far increased likelihood of inadvertent bugs: Because of the broken "viral" nature of unsigned quantities in C (operations involving mixed signed/unsigned are done unsigned, when exactly the opposite is nearly always wanted), even a single error in declaring a quantity unsigned that should be signed, or even the even more subtle error of comparing signed and unsigned values and forgetting the necessary cast, can be catastrophic, as comparisons will yield wrong results. -Wsign-compare is turned on specifically to catch this, but this tends to result in a great number of warnings when mixing signed and unsigned, and the casts are annoying. More has been written on this elsewhere. -- All such quantity types just mentioned boil down to EMACS_INT, which is 32 bits on 32-bit machines and 64 bits on 64-bit machines. This is guaranteed to be the same size as Lisp objects of type `int', and (as far as I can tell) of size_t (unsigned!) and ssize_t. The only type below that is not an EMACS_INT is Hashcode, which is an unsigned value of the same size as EMACS_INT. -- Type names should be relatively short (no more than 10 characters or so), with the first letter capitalized and no underscores if they can at all be avoided. -- "count" == a zero-based measurement of some quantity. Includes sizes, offsets, and indexes. -- "bpos" == a one-based measurement of a position in a buffer. "Charbpos" and "Bytebpos" count text in the buffer, rather than bytes in memory; thus Bytebpos does not directly correspond to the memory representation. Use "Membpos" for this. -- "Char" refers to internal-format characters, not to the C type "char", which is really a byte. -- For the actual name changes, see the script below. I ran the following script to do the conversion. (NOTE: This script is idempotent. You can safely run it multiple times and it will not screw up previous results -- in fact, it will do nothing if nothing has changed. Thus, it can be run repeatedly as necessary to handle patches coming in from old workspaces, or old branches.) There are two tags, just before and just after the change: `pre-integral-type-rename' and `post-integral-type-rename'. When merging code from the main trunk into a branch, the best thing to do is first merge up to `pre-integral-type-rename', then apply the script and associated changes, then merge from `post-integral-type-change' to the present. (Alternatively, just do the merging in one operation; but you may then have a lot of conflicts needing to be resolved by hand.) Script `fixtypes.sh' follows: ----------------------------------- cut ------------------------------------ files="*.[ch] s/*.h m/*.h config.h.in ../configure.in Makefile.in.in ../lib-src/*.[ch] ../lwlib/*.[ch]" gr Memory_Count Bytecount $files gr Lstream_Data_Count Bytecount $files gr Element_Count Elemcount $files gr Hash_Code Hashcode $files gr extcount bytecount $files gr bufpos charbpos $files gr bytind bytebpos $files gr memind membpos $files gr bufbyte intbyte $files gr Extcount Bytecount $files gr Bufpos Charbpos $files gr Bytind Bytebpos $files gr Memind Membpos $files gr Bufbyte Intbyte $files gr EXTCOUNT BYTECOUNT $files gr BUFPOS CHARBPOS $files gr BYTIND BYTEBPOS $files gr MEMIND MEMBPOS $files gr BUFBYTE INTBYTE $files gr MEMORY_COUNT BYTECOUNT $files gr LSTREAM_DATA_COUNT BYTECOUNT $files gr ELEMENT_COUNT ELEMCOUNT $files gr HASH_CODE HASHCODE $files ----------------------------------- cut ------------------------------------ `fixtypes.sh' is a Bourne-shell script; it uses 'gr': ----------------------------------- cut ------------------------------------ #!/bin/sh # Usage is like this: # gr FROM TO FILES ... # globally replace FROM with TO in FILES. FROM and TO are regular expressions. # backup files are stored in the `backup' directory. from="$1" to="$2" shift 2 echo ${1+"$@"} | xargs global-replace "s/$from/$to/g" ----------------------------------- cut ------------------------------------ `gr' in turn uses a Perl script to do its real work, `global-replace', which follows: ----------------------------------- cut ------------------------------------ : #-*- Perl -*- ### global-modify --- modify the contents of a file by a Perl expression ## Copyright (C) 1999 Martin Buchholz. ## Copyright (C) 2001 Ben Wing. ## Authors: Martin Buchholz <martin@xemacs.org>, Ben Wing <ben@xemacs.org> ## Maintainer: Ben Wing <ben@xemacs.org> ## Current Version: 1.0, May 5, 2001 # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2, or (at your option) # any later version. # # This program is distributed in the hope that it will be useful, but # WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU # General Public License for more details. # # You should have received a copy of the GNU General Public License # along with XEmacs; see the file COPYING. If not, write to the Free # Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA # 02111-1307, USA. eval 'exec perl -w -S $0 ${1+"$@"}' if 0; use strict; use FileHandle; use Carp; use Getopt::Long; use File::Basename; (my $myName = $0) =~ s@.*/@@; my $usage=" Usage: $myName [--help] [--backup-dir=DIR] [--line-mode] [--hunk-mode] PERLEXPR FILE ... Globally modify a file, either line by line or in one big hunk. Typical usage is like this: [with GNU print, GNU xargs: guaranteed to handle spaces, quotes, etc. in file names] find . -name '*.[ch]' -print0 | xargs -0 $0 's/\bCONST\b/const/g'\n [with non-GNU print, xargs] find . -name '*.[ch]' -print | xargs $0 's/\bCONST\b/const/g'\n The file is read in, either line by line (with --line-mode specified) or in one big hunk (with --hunk-mode specified; it's the default), and the Perl expression is then evalled with \$_ set to the line or hunk of text, including the terminating newline if there is one. It should destructively modify the value there, storing the changed result in \$_. Files in which any modifications are made are backed up to the directory specified using --backup-dir, or to `backup' by default. To disable this, use --backup-dir= with no argument. Hunk mode is the default because it is MUCH MUCH faster than line-by-line. Use line-by-line only when it matters, e.g. you want to do a replacement only once per line (the default without the `g' argument). Conversely, when using hunk mode, *ALWAYS* use `g'; otherwise, you will only make one replacement in the entire file! "; my %options = (); $Getopt::Long::ignorecase = 0; &GetOptions ( \%options, 'help', 'backup-dir=s', 'line-mode', 'hunk-mode', ); die $usage if $options{"help"} or @ARGV <= 1; my $code = shift; die $usage if grep (-d || ! -w, @ARGV); sub SafeOpen { open ((my $fh = new FileHandle), $_[0]); confess "Can't open $_[0]: $!" if ! defined $fh; return $fh; } sub SafeClose { close $_[0] or confess "Can't close $_[0]: $!"; } sub FileContents { my $fh = SafeOpen ("< $_[0]"); my $olddollarslash = $/; local $/ = undef; my $contents = <$fh>; $/ = $olddollarslash; return $contents; } sub WriteStringToFile { my $fh = SafeOpen ("> $_[0]"); binmode $fh; print $fh $_[1] or confess "$_[0]: $!\n"; SafeClose $fh; } foreach my $file (@ARGV) { my $changed_p = 0; my $new_contents = ""; if ($options{"line-mode"}) { my $fh = SafeOpen $file; while (<$fh>) { my $save_line = $_; eval $code; $changed_p = 1 if $save_line ne $_; $new_contents .= $_; } } else { my $orig_contents = $_ = FileContents $file; eval $code; if ($_ ne $orig_contents) { $changed_p = 1; $new_contents = $_; } } if ($changed_p) { my $backdir = $options{"backup-dir"}; $backdir = "backup" if !defined ($backdir); if ($backdir) { my ($name, $path, $suffix) = fileparse ($file, ""); my $backfulldir = $path . $backdir; my $backfile = "$backfulldir/$name"; mkdir $backfulldir, 0755 unless -d $backfulldir; print "modifying $file (original saved in $backfile)\n"; rename $file, $backfile; } WriteStringToFile ($file, $new_contents); } } ----------------------------------- cut ------------------------------------ In addition to those programs, I needed to fix up a few other things, particularly relating to the duplicate definitions of types, now that some types merged with others. Specifically: 1. in lisp.h, removed duplicate declarations of Bytecount. The changed code should now look like this: (In each code snippet below, the first and last lines are the same as the original, as are all lines outside of those lines. That allows you to locate the section to be replaced, and replace the stuff in that section, verifying that there isn't anything new added that would need to be kept.) --------------------------------- snip ------------------------------------- /* Counts of bytes or chars */ typedef EMACS_INT Bytecount; typedef EMACS_INT Charcount; /* Counts of elements */ typedef EMACS_INT Elemcount; /* Hash codes */ typedef unsigned long Hashcode; /* ------------------------ dynamic arrays ------------------- */ --------------------------------- snip ------------------------------------- 2. in lstream.h, removed duplicate declaration of Bytecount. Rewrote the comment about this type. The changed code should now look like this: --------------------------------- snip ------------------------------------- #endif /* The have been some arguments over the what the type should be that specifies a count of bytes in a data block to be written out or read in, using Lstream_read(), Lstream_write(), and related functions. Originally it was long, which worked fine; Martin "corrected" these to size_t and ssize_t on the grounds that this is theoretically cleaner and is in keeping with the C standards. Unfortunately, this practice is horribly error-prone due to design flaws in the way that mixed signed/unsigned arithmetic happens. In fact, by doing this change, Martin introduced a subtle but fatal error that caused the operation of sending large mail messages to the SMTP server under Windows to fail. By putting all values back to be signed, avoiding any signed/unsigned mixing, the bug immediately went away. The type then in use was Lstream_Data_Count, so that it be reverted cleanly if a vote came to that. Now it is Bytecount. Some earlier comments about why the type must be signed: This MUST BE SIGNED, since it also is used in functions that return the number of bytes actually read to or written from in an operation, and these functions can return -1 to signal error. Note that the standard Unix read() and write() functions define the count going in as a size_t, which is UNSIGNED, and the count going out as an ssize_t, which is SIGNED. This is a horrible design flaw. Not only is it highly likely to lead to logic errors when a -1 gets interpreted as a large positive number, but operations are bound to fail in all sorts of horrible ways when a number in the upper-half of the size_t range is passed in -- this number is unrepresentable as an ssize_t, so code that checks to see how many bytes are actually written (which is mandatory if you are dealing with certain types of devices) will get completely screwed up. --ben */ typedef enum lstream_buffering --------------------------------- snip ------------------------------------- 3. in dumper.c, there are four places, all inside of switch() statements, where XD_BYTECOUNT appears twice as a case tag. In each case, the two case blocks contain identical code, and you should *REMOVE THE SECOND* and leave the first.

;;; paths.el --- define pathnames for use by various Emacs commands.

;; Copyright (C) 1986, 1988, 1993, 1994, 1997 Free Software Foundation, Inc.

;; Maintainer: FSF
;; Keywords: internal, dumped

;; This file is part of XEmacs.

;; XEmacs is free software; you can redistribute it and/or modify it
;; under the terms of the GNU General Public License as published by
;; the Free Software Foundation; either version 2, or (at your option)
;; any later version.

;; XEmacs is distributed in the hope that it will be useful, but
;; WITHOUT ANY WARRANTY; without even the implied warranty of
;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
;; General Public License for more details.

;; You should have received a copy of the GNU General Public License
;; along with XEmacs; see the file COPYING.  If not, write to the Free
;; Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.

;;; Synched up with: FSF 19.30.

;;; Commentary:

;; This file is dumped with XEmacs.

;; These are default settings for names of certain files and directories
;; that Emacs needs to refer to from time to time.

;; If these settings are not right, override them with `setq'
;; in site-start.el.  Do not change this file.

;;; Code:

;Note: FSF's version is:
;(defvar Info-default-directory-list
;  (let ((start (list "/usr/local/lib/info/"
;		      ;; This comes second so that, if it is the same
;		      ;; as configure-info-directory (which is usually true)
;		      ;; and Emacs has been installed (also usually true)
;		      ;; then the list will end with two copies of this;
;		      ;; which means that the last dir file Info-insert-dir
;		      ;; finds will be the one in this directory.
;		      "/usr/local/info/"))
;	 (configdir (file-name-as-directory configure-info-directory)))
;    (setq start (nconc start (list configdir)))
;    start)
;  "List of directories to search for Info documentation files.
;They are searched in the order they are given in this list.
;Therefore, the directory of Info files that come with Emacs
;normally should come last (so that local files override standard ones).")

;Our commented-out version is:
;(defvar Info-default-directory-list
;  (let ((start (list "/usr/local/info/"
;		     "/usr/local/lib/info/"))
;	(configdir (file-name-as-directory configure-info-directory)))
;    (or (member configdir start)
;	(setq start (nconc start (list configdir))))
;    (or (member (expand-file-name "../info/" data-directory) start)
;	(setq start
;	      (nconc start
;		     (list (expand-file-name "../info/" data-directory)))))
;    start)
;  "List of directories to search for Info documentation files.")

(defvar news-path "/usr/spool/news/"
  "The root directory below which all news files are stored.")

(defvar news-inews-program nil
  "Program to post news.")

;(defvar gnus-default-nntp-server ""
;  ;; set this to your local server
;  "The name of the host running an NNTP server.
;If it is a string such as \":DIRECTORY\", then ~/DIRECTORY
;is used as a news spool.  `gnus-nntp-server' is initialized from NNTPSERVER
;environment variable or, if none, this value.")

;(defvar gnus-nntp-service "nntp"
;  "NNTP service name, usually \"nntp\" or 119).
;Go to a local news spool if its value is nil, in which case `gnus-nntp-server'
;should be set to `(system-name)'.")

(defvar mh-progs nil
  "Directory containing MH commands.")

(defvar mh-lib nil
  "Directory of MH library.")

(defvar rmail-file-name "~/RMAIL"
  "Name of user's primary mail file.")

(defconst rmail-spool-directory nil
  "Name of directory used by system mailer for delivering new mail.
Its name should end with a slash.")

(defconst sendmail-program nil
  "Program used to send messages.")

(defconst remote-shell-program nil
  "Program used to execute shell commands on a remote machine.")

(defconst term-file-prefix "term/"
  "If non-nil, Emacs startup does (load (concat term-file-prefix (getenv \"TERM\")))
You may set this variable to nil in your `.emacs' file if you do not wish
the terminal-initialization file to be loaded.")

(defconst manual-program nil
  "Program to run to print man pages.")

(defconst abbrev-file-name "~/.abbrev_defs"
  "*Default name of file to read abbrevs from.")

(defconst directory-abbrev-alist nil)

;; Formerly, the values of these variables were computed once
;; (at dump time).  However, with the advent of pre-compiled binaries
;; and homebrewed systems such as Linux where who knows where the
;; hell the various programs may be located (if they even exist at all),
;; it's clear that we need to recompute these values at run time.
;; In typical short-sightedness, site administrators have been told up
;; till now to do `setq's in site-init.el, which is run only once --
;; at dump time.  So we have to do contortions to make sure we don't
;; override values set in site-init.el.

(defun initialize-xemacs-paths ()
  "Initialize the XEmacs path variables from the environment.
Called automatically at dump time and run time.  Do not call this.
Will not override settings in site-init.el or site-run.el."
  (let ((l #'(lambda (var value)
	       (let ((origsym (intern (concat "paths-el-original-"
					      (symbol-name var)))))
		 (if (running-temacs-p)
		     (progn
		       (set var value)
		       (set origsym value))
		   (and (eq (symbol-value var) (symbol-value origsym))
			(set var value)))))))
    (funcall
     l 'news-inews-program
     (cond ((file-exists-p "/usr/bin/inews") "/usr/bin/inews")
	   ((file-exists-p "/usr/local/inews") "/usr/local/inews")
	   ((file-exists-p "/usr/local/bin/inews") "/usr/local/bin/inews")
	   ((file-exists-p "/usr/lib/news/inews") "/usr/lib/news/inews")
	   (t "inews")))

    (funcall
     l 'mh-progs
     (cond ((file-directory-p "/usr/bin/mh") "/usr/bin/mh/") ;Ultrix 4.2
	   ((file-directory-p "/usr/new/mh") "/usr/new/mh/") ;Ultrix <4.2
	   ((file-directory-p "/usr/local/bin/mh") "/usr/local/bin/mh/")
	   ((file-directory-p "/usr/local/mh") "/usr/local/mh/")
	   (t "/usr/local/bin/")))

    (funcall
     l 'mh-libs
     (cond ((file-directory-p "/usr/lib/mh") "/usr/lib/mh/") ;Ultrix 4.2
	   ((file-directory-p "/usr/new/lib/mh")
	    "/usr/new/lib/mh/") ;Ultrix <4.2
	   ((file-directory-p "/usr/local/lib/mh") "/usr/local/lib/mh/")
	   (t "/usr/local/bin/mh/")))

    (funcall
     l 'rmail-spool-directory
     (cond ((string-match "^[^-]+-[^-]+-sco3.2v4" system-configuration)
	    "/usr/spool/mail/")
	   ;; On The Bull DPX/2 /usr/spool/mail is used although
	   ;; it is usg-unix-v.
	   ((string-match "^m68k-bull-sysv3" system-configuration)
	    "/usr/spool/mail/")
	   ;; SVR4 and recent BSD are said to use this.
	   ;; Rather than trying to know precisely which systems use it,
	   ;; let's assume this dir is never used for anything else.
	   ((file-exists-p "/var/mail")
	    "/var/mail/")
	   ((memq system-type '(dgux hpux usg-unix-v unisoft-unix rtu irix))
	    "/usr/mail/")
	   ((memq system-type '(linux))
	    "/var/spool/mail/")
	   (t "/usr/spool/mail/")))

    (funcall
     l 'sendmail-program
     (cond
      ((file-exists-p "/usr/lib/sendmail") "/usr/lib/sendmail")
      ((file-exists-p "/usr/sbin/sendmail") "/usr/sbin/sendmail")
      ((file-exists-p "/usr/ucblib/sendmail") "/usr/ucblib/sendmail")
      (t "fakemail")))		;In ../etc, to interface to /bin/mail.

    (funcall
     l 'remote-shell-program
     (cond
      ;; Some systems use rsh for the remote shell; others use that
      ;; name for the restricted shell and use remsh for the remote
      ;; shell.  Let's try to guess based on what we actually find
      ;; out there.  The restricted shell is almost certainly in
      ;; /bin or /usr/bin, so it's probably safe to assume that an
      ;; rsh found elsewhere is the remote shell program.  The
      ;; converse is not true: /usr/bin/rsh could be either one, so
      ;; check that last.
      ((file-exists-p "/usr/ucb/remsh") "/usr/ucb/remsh")
      ((file-exists-p "/usr/bsd/remsh") "/usr/bsd/remsh")
      ((file-exists-p "/bin/remsh") "/bin/remsh")
      ((file-exists-p "/usr/bin/remsh") "/usr/bin/remsh")
      ((file-exists-p "/usr/local/bin/remsh") "/usr/local/bin/remsh")
      ((file-exists-p "/usr/ucb/rsh") "/usr/ucb/rsh")
      ((file-exists-p "/usr/bsd/rsh") "/usr/bsd/rsh")
      ((file-exists-p "/usr/local/bin/rsh") "/usr/local/bin/rsh")
      ((file-exists-p "/usr/bin/rcmd") "/usr/bin/rcmd")
      ((file-exists-p "/bin/rcmd") "/bin/rcmd")
      ((file-exists-p "/bin/rsh") "/bin/rsh")
      ((file-exists-p "/usr/bin/rsh") "/usr/bin/rsh")
      (t "rsh")))

    (funcall
     l 'manual-program
     ;; Solaris 2 has both of these files; prefer /usr/ucb/man
     ;; because the other has nonstandard argument conventions.
     (if (file-exists-p "/usr/ucb/man")
	 "/usr/ucb/man" "/usr/bin/man"))

    (funcall
     l 'directory-abbrev-alist
     ;; Try to match various conventions for automounter temporary
     ;; mount points.  These temporary mount points may go away, so
     ;; it's important that we only try to read files under the
     ;; "advertised" mount point, rather than the temporary one, or it
     ;; will look like files have been deleted on us.  Whoever came up
     ;; with this design is clearly a moron of the first order, but
     ;; now we're stuck with it, no doubt until the end of time.
     ;;
     ;; For best results, automounter junk should go near the front of this
     ;; list, and other user translations should come after it.
     ;;
     ;; Our code handles the following empirically observed conventions:
     ;; /net is an actual directory! (some systems are not broken!)
     ;; /net/HOST -> /tmp_mnt/net/HOST (`standard' old Sun automounter)
     ;; /net/HOST -> /tmp_mnt/HOST (BSDI 4.0)
     ;; /net/HOST -> /a/HOST (Freebsd 2.2.x)
     ;; /net/HOST -> /amd/HOST (seen in amd sample config files)
     ;;
     ;; If your system has a different convention, you may have to change this.
     ;; Don't forget to send in a patch!
     (when (file-directory-p "/net")
       (append
	(when (file-directory-p "/tmp_mnt")
	  (if (file-directory-p "/tmp_mnt/net")
	      '(("\\`/tmp_mnt/net/" . "/net/"))
	    '(("\\`/tmp_mnt/" . "/net/"))))
	(when (file-directory-p "/a")
	  '(("\\`/a/" . "/net/")))
	(when (file-directory-p "/amd")
	  '(("\\`/amd/" . "/net/")))
	)))
))

(if (running-temacs-p)
    (initialize-xemacs-paths))

;;; paths.el ends here